Periodic Reporting for period 1 - V4SAFETY (Vehicles and VRU Virtual eValuation of Road Safety)
Période du rapport: 2022-10-01 au 2024-03-31
An accepted, reliable method for the evaluation and comparison of safety measures for Connected, Cooperative and Automated Mobility is needed in order to set policies for road safety in the coming decades and to push for Vision Zero. The V4SAFETY method will deal with the safety of all road users, from vulnerable road users to vehicle occupants.
V4SAFETY will provide a prospective safety assessment framework that can handle a large variety of safety measures, ranging from in-vehicle safety technology, new vehicle types, infrastructure solutions to regulations that influence road user behaviour. It includes methods to project the results onto future scenarios and across EU regions for use by policy makers, authorities, and consumer organizations.
The method provides tools to characterise the influence of the contributing factors and their uncertainties, as well as guidelines for its application, leading to greater understanding of the differences between prospective assessment studies and of the influence of underlying data, assumptions, and models. The resulting transparency and consistency in simulation-based safety assessment leads to much improved comparability and reliability of assessment conclusions.
The V4SAFETY consortium, led by TNO, consists of following international partners:
• Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek (TNO)
• Bayerische Motoren Werke AG (BMW)
• Bundesanstalt für Strassenwesen (BASt)
• Chalmers Tekniska Hogskola AB (CHAL)
• Fraunhofer Gesellschaft zur Förderung der angewandten Forschung EV (IVI)
• IDIADA Automotive Technology SA (IDI)
• Rheinisch-Westfälische Technische Hochschule (ika)
• Groupement D’interet Economique de Recherches Et d’etudes PSA Renault (LAB)
• Stichting Wetenschappelijk Onderzoek Verkeersveiligheid (SWOV)
• Technische Hochschule Ingolstadt (THI)
• Toyota Motor Europe NV (TME)
• Universita Degli Studi Di Firenze (UNIFI)
• Virtual Vehicle Research GmbH (VIF)
• Volvo Personvagnar AB (VCC)
• ZF Friedrichshafen AG (ZF)
• European Road Transport Telematics Implementation Coordination Organisation – Intelligent Transport Systems & Services Europe (ERTICO)
• W2Economics (W2E)
The project will run for 3 years (October 2022 – September 2025).
V4SAFETY will provide a prospective safety assessment framework that can handle a large variety of safety measures, ranging from in-vehicle safety technology, new vehicle types, infrastructure solutions to regulations that influence road user behaviour. It includes methods to project the results onto future scenarios and across EU regions for use by policy makers, authorities, and consumer organizations.
The method provides tools to characterise the influence of the contributing factors and their uncertainties, as well as guidelines for its application, leading to greater understanding of the differences between prospective assessment studies and of the influence of underlying data, assumptions, and models. The resulting transparency and consistency in simulation-based safety assessment leads to much improved comparability and reliability of assessment conclusions.
The V4SAFETY consortium, led by TNO, consists of following international partners:
• Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek (TNO)
• Bayerische Motoren Werke AG (BMW)
• Bundesanstalt für Strassenwesen (BASt)
• Chalmers Tekniska Hogskola AB (CHAL)
• Fraunhofer Gesellschaft zur Förderung der angewandten Forschung EV (IVI)
• IDIADA Automotive Technology SA (IDI)
• Rheinisch-Westfälische Technische Hochschule (ika)
• Groupement D’interet Economique de Recherches Et d’etudes PSA Renault (LAB)
• Stichting Wetenschappelijk Onderzoek Verkeersveiligheid (SWOV)
• Technische Hochschule Ingolstadt (THI)
• Toyota Motor Europe NV (TME)
• Universita Degli Studi Di Firenze (UNIFI)
• Virtual Vehicle Research GmbH (VIF)
• Volvo Personvagnar AB (VCC)
• ZF Friedrichshafen AG (ZF)
• European Road Transport Telematics Implementation Coordination Organisation – Intelligent Transport Systems & Services Europe (ERTICO)
• W2Economics (W2E)
The project will run for 3 years (October 2022 – September 2025).
The V4SAFETY project has produced a draft Safety Assessment Framework (Milestone 10, Oct 2023) that gives a first description of the full process. This already gives guidance for users on how to set up a comprehensive prospective safety assessment.
A safety assessment needs a baseline situation to compare to. D4.1 (March 2024) characterises the various data sources, provides requirements for data, provides dimensions and metrics for grading and recommendations for baseline data.
Other milestones in the project were (status March 2024):
• MS3: Draft literature review across all human models
• MS4: Final List of Use Cases and Safety Measures for Demonstration
• MS11: Draft recommendations for simulation structure
• MS12: Baseline Data for Demonstration
• MS14: Provision of draft traffic conflict-related behaviour models
• MS15: Draft Method Overview for Baseline Generation
A safety assessment needs a baseline situation to compare to. D4.1 (March 2024) characterises the various data sources, provides requirements for data, provides dimensions and metrics for grading and recommendations for baseline data.
Other milestones in the project were (status March 2024):
• MS3: Draft literature review across all human models
• MS4: Final List of Use Cases and Safety Measures for Demonstration
• MS11: Draft recommendations for simulation structure
• MS12: Baseline Data for Demonstration
• MS14: Provision of draft traffic conflict-related behaviour models
• MS15: Draft Method Overview for Baseline Generation
V4SAFETY Safety Assessment Framework: most comprehensive framework in the field, including both vehicle, infrastructure and policy-oriented safety measures. Including a first validation approach to validate the assessment.
Human in simulation:
• Elaborate literature review of road-user behaviour models.
• A probabilistic driver behaviour model on drivers’ responses to Forward Collision Warnings has been specified for WP6 partners implementation and use in simulation.
• Partners have further developed and adapted their existing computational behaviour models to the V4SAFETY use-case scenarios.
• We have started to document (in D3.1) challenges and recommendations for the development, validation and use of the different types of computational behaviour models identified for virtual safety assessment.
Baseline:
• Method to define a baseline based on the approaches defined in the framework, representation methods, data processing, test scenario definition, pre-simulation variants and baseline requirements for in-simulation models (to be included in D4.2)
• D4.1 Data Sources for Baseline Generation – Overview, Grading, and Recommendations
Modelling safety measures:
• First guidelines on defining a simulation structure were defined.
• Modelling approaches for pre-and in-crash simulation elements and interfaces were collected.
• An AEB model was created and provided for in-and external usage.
Human in simulation:
• Elaborate literature review of road-user behaviour models.
• A probabilistic driver behaviour model on drivers’ responses to Forward Collision Warnings has been specified for WP6 partners implementation and use in simulation.
• Partners have further developed and adapted their existing computational behaviour models to the V4SAFETY use-case scenarios.
• We have started to document (in D3.1) challenges and recommendations for the development, validation and use of the different types of computational behaviour models identified for virtual safety assessment.
Baseline:
• Method to define a baseline based on the approaches defined in the framework, representation methods, data processing, test scenario definition, pre-simulation variants and baseline requirements for in-simulation models (to be included in D4.2)
• D4.1 Data Sources for Baseline Generation – Overview, Grading, and Recommendations
Modelling safety measures:
• First guidelines on defining a simulation structure were defined.
• Modelling approaches for pre-and in-crash simulation elements and interfaces were collected.
• An AEB model was created and provided for in-and external usage.